Creating a Cluster

Note: To create a cluster with pgEdge Cloud, you must have a pgEdge Enterprise Edition account (opens in a new tab) and a linked Cloud provider account.

To create a cluster, select the + New Cluster button in the left navigation pane. Then:

• Provide a unique name for your cluster in the Cluster Name field.
• Use the drop-down listbox in the Cloud Account field to select the cloud provider account that will be used to provision your cluster.

When the Create New Cluster dialog updates, select the regions in which you want to deploy nodes:

• Click an icon on the map or in the Regions field to add a node in the selected region; when selected, the name of the region is added to the list of selectedRegions. To deselect a region, click the icon again, or use the X in the listed region's name icon to remove it.

AWS Users: Note that while you can access all regions in the pgEdge console, not all regions shown may be enabled for use in your AWS account. To enable a region, see Enabling a Region in the AWS Console.

Each node of your cluster is represented by a pane in the Network Settings section that displays the node addresses. Toggles at the top of the section control the display:

• Use the Configuration toggle to accept the Default address assignments, or move the toggle to Manual to specify custom address assignments.
• Use the Network Type toggle to specify if the connection values displayed or edited should be Public or Private addresses.

You can assign network addresses that work with network address ranges (CIDR ranges) currently in use on your private network; all addresses should meet Class C address specifications (opens in a new tab).

• Use the Network address for outside connections from client applications, SSH requests, and connections from web clients.
• Use the Public Subnet and Private Subnet addresses for VPC peering within your own network.

Use the fields in the Configure Your Cloud Environment section to specify the initial configuration of your cluster; note that the cloud environment options will vary with the provider chosen:

• Use the Instance Type drop-down to select the size and configuration of your cluster instance type. Each node in the cluster will reside on the instance type selected. The instance type is one factor in determining the operating cost of your cluster.

• Use the Volume Size (GB) field to specify the size of the data volume for each node of your cluster. Volume size must be between 8 GB and 10 TB.

• Use the SSH Key drop-down to select a key pair from the list of your imported key pairs. A key pair is required to SSH to a cluster; without a key, you can connect to databases you provision on the cluster, but will not be able to connect with the host via SSH.

Use fields in the Configure Your Backup Storage section to select the backup stores (opens in a new tab) that will be available to use for database backups and log file backups for the new cluster.

As you select backup stores, remember that selecting backup stores near your cluster nodes will reduce network latency during the backup process. Use the drop-down field or select locations on the map to include a specific store.

Use the fields in the Add Firewall Rules section to open ports on each node of your cluster for connections.

• Use the Type drop-down to select the connection type you are defining:

  • Select PostgreSQL to create a connection to port 5432 ; this is the port that connecting clients that wish to query the database should use.

  • Select HTTPS to open port 443 for secure communication between PostgreSQL and a web server.

  • Select SSH to open port 22 for SSH connections.

• Use the Sources drop-down to select the IP address(es) that will be allowed to connect to the port.

  • Choose My IP:x.x.x.x to open a port for the detected IP address from which you are connected.
  • Choose All IPV4 to allow connections to the selected port from any source. This setting is not recommended for connections to the database server.

After selecting a protocol type and an address, choose the + Add Rule button to add the list of firewall rules. To remove a rule from the list, select the X in the upper-right corner of the rule box.

To create the cluster without firewall rules, check the box to the left of Deploy cluster without firewall rules. Note that if you do not open ports for connection, client software will be unable to connect to the OS or databases that reside on the cluster.

In the Select Additional Options section you'll find the option to create the cluster as a single-database cluster.

Single database clusters are beneficial because:

• they allow connections to the database with an IP address (rather than by domain name).
• they allow connections to the database via DNS names and records that you create outside of the pgEdge console.
• a single-purpose, dedicated cluster guarantees that all cluster resources are available for the single database.

When you've completed the dialog, select the Deploy Cluster button to deploy a provisioned cluster.

A progress bar tracks the progress as the deployment goes through each step; when the cluster is ready to use, it is added to the list of clusters in the navigation tree.

If you encounter any errors during cluster creation, please see the Troubleshooting section.